Broad binding-site specificity and affinity properties of octamer 1 and brain octamer-binding proteins

Broad binding-site specificity and affinity properties of octamer 1 and brain octamer-binding proteins
of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Eur. J. Biochem. zyxwvusrqpo 17, 799-811 (1993) zyxwvu   EBS 1993 zyxwvusrq Broad binding-site specificity and affinity properties of octamer 1 and brain octamer-binding proteins Andrew J. BENDALL'.2, Richard A. STURM3, Patrick A. C. DANOY' and Peter L. MOLLOY' zyx   chool of Biological Sciences, Macquarie University, New South Wales, Australia Commonwealth Scientific and Industrial Research Organisation Division of Biomolecular Engineering, New South Wales, Australia Centre for Molecular Biology and Biotechnology, University of Queensland, Australia (Received June 1WAugust 10, 1993) EJB 93 0881/1 The ubiquitous Pit-1-Oct-l-Unc-1 (P0U)-domain protein octamer 1 (Oct-1) has been observed to bind specifically to a number of degenerate and dissimilar sequences. We have used antibodies directed against a C-terminal Oct-1 peptide to immunoselect binding sequences for HeLa cell Oct-1 from random-sequence oligonucleotides and we describe the isolation of binding sequences of considerable heterogeneity. Although our consensus alignment indicated a 9-bp TATGCAAAT motif with AT-rich flanking sequences, this binding motif is not immediately obvious in the popula- tion of sequences and no clone actually contained this sequence. Screening these Oct-l-binding sequences with a mouse whole-brain extract demonstrated that the neuronal octamer-binding pro- teins exhibit similar but distinct DNA sequence specificities. Unlike the reported selection of binding sequences for other transcription factors, the dependence of Oct-I -binding affinity upon sequence did not correspond tightly to the degree of conservation at particular positions of the consensus sequence. Our results suggest that either base-specific hydrogen bonding is not the only major determinant of binding affinity and specificity, or that Oct-1 binding to some sequences is mecha- nistically different from its binding to an octamer. These results exemplify the potential to overlook binding sites for some factors by searching gene sequences with a consensus nucleotide sequence. Sequence-specific recognition by DNA-binding tran- scription factors is a central mechanism in the control of gene expression. This property has been widely used in the identi- fication of both gene-regulatory proteins and the conserved motifs that are their target binding sites in regions of tran- scriptional control. However, it is now apparent that se- quence recognition by a number of transcription factors is more relaxed than previously recognised and many target sites may not be readily identified using a consensus derived from an ad hoc collection of binding sites. The ubiquitously expressed protein octamer 1 (Oct-1) is a transcription factor having remarkably flexible DNA-sequence recognition prop- erties. Oct-1 has been found to be involved in stimulating RNA-polymerase-I1 transcription of cellular genes that are constitutively active (Mattaj et al., 1985; Ares et al., 1987), cell-cycle regulated (LaBella et al., 1988), tissue-specific (LeBowitz et al., 1988; Johnson et al., 1990; Kemler et al., 1991) or developmentally regulated (Zwartkruis et al., 1992). It has also been shown to be involved in RNA-polymerase- zyxwv orrespondence to P. L. Molloy, Commonwealth Scientific and Industrial Research Organisation Division of Biomolecular Engi- neering, Sydney Laboratory, PO Box 184, North Ryde, New South Wales, Australia 2113 Fax: + 61 2 886 4805. Abbreviations. EMSA, electrophoretic mobility-shift assay; HSV, herpes simplex virus ; KLH, keyhole-limpet haemocyanin ; MLP, ad- enovirus major late promoter; PCR, polymerase chain reaction; PhMeSO,F, phenylmethylsulphonyl fluoride; SV40 simian virus 40; Oct, octamer; POU, Pit-l-Oct-l-Unc-l ; ds, double-stranded. 111 transcription (Bark et al., 1987; Carbon et al., 1987) and the stimulation of viral-DNA replication (Pruijn et al., 1986; Pruijn et al., 1987; Verrijzer et al., 1990a). The participation of Oct-1 at these regulatory elements is dependent on the presence of a highly conserved motif that has been defined as an octamer, ATGCAAAT (Parslow et al., 1984), or as a decamer, ATGCAAATNA (Falkner and Zachau, 1984; Falk- ner et al., 1986), that represents a core binding site for this protein. However, a number of degenerate octamer motifs that represent Oct-l-binding sites occur in viral-transcription- al-regulatory regions such as the simian virus 40 (SV40) en- hancer (Stunn et al., 1987), mouse-mammary-tumour-virus promoter (Briiggemeier et al., 1991), human-papillomavirus enhancers (Chong et al., 1991 ; Hoppe-Seyler et al., 1991) and the adenovirus inverted terminal repeats that function as srcins of replication (Kelly, 1984; Pruijn et al., 1988). It is further known that Oct-1 can independently bind to other conserved motifs like the heptamer CTCATGA, found along- side the octamer in immunoglobulin heavy-chain promoters (Landolfi et al., 1988; Kemler et al., 1989; Poellinger and Roeder, 1989), or the TAATGARAT motif of herpes-sim- plex-virus (HSV) immediate early genes (Mackem and Roiz- man, 1982; O'Hare and Goding, 1988; ap Rhys et al., 1989). Sequences containing ATAAAT in the human c-myc and Ay- globin promoters (Takimoto et al., 1989; Ponce et al., 1991) are also targets for the Oct-1 protein. The DNA-binding domain of Oct-1 is referred to as the Pit-l-Oct-1 -Unc-1 (POU) domain and is conserved among a family of factors that are known or thought to be involved  800 in the regulation of transcription (Rosenfeld, 1991 zyxwvu   Scholer, 1991). The POU domain consists of about 160 amino acids that can be subdivided into two regions designated as the POU-specific domain of 75-82 amino acids, and the POU homeo domain of 60 amino acids, connected by a short linker of variable length and sequence (Herr et al., 1988; Sturm and Herr, 1988). Computer modelling suggests that the POU homeo domain can adopt a tri-helical conformation analo- gous to the DNA-binding motifs encoded by developmen- tally important homeo box genes (Kissinger et al., 1990; Otting et al., 1990) and related motifs in prokaryotic repres- sors (Laughon and Scott, 1984: Gehring et al., 1990). The recognition helix of this tri-helical bundle is thought to lie within the major groove of the AT-rich 3' end of the octamer (Kristie and Sharp, 1990: Verrijzer et al., 1990b). The POU- specific domain has been shown to interact and make specific contacts with the 5'-end ATGC of the octamer (Kristie and Sharp, 1990: Verrijzer et al., 1990b). While no recognisable DNA binding structure was initially identified in the POU- specific domain, it has recently been demonstrated to consist of four a-helices, two of which adopt a similar structure to the bacterial helix-turn-helix motif (Dekker et al., 1993 ; Assa-Munt et al., 1993). This bipartite structure of two DNA- binding subdomains flexibly linked in the same polypeptide (Sturm and Herr, 1988) is unusual amongst transcription factors and may offer a structural basis for the flexibility of DNA binding by Oct-1. Chemical interference analysis showed that modification of all but one purine in the double-stranded (ds) ATGCAAAT core sequence disrupted binding (Baumruker et al., 1988). This suggested that the overall affinity of Oct-1 for a se- quence represents the aggregate of many specific base in- teractions though the indirect effect that base modification has on helix geometry was considered (Baumruker et al., 1988). Moreover, significant nucleotide and backbone con- tacts of full length Oct-1 or its POU domain have been shown to extend to the zyxwvutsr ' side of the octamer motif (Sturm et al., 1987; Baumruker et al., 1988; Pruijn et al., 1988; Verrijzer et al., 1990b), suggesting that bases 5' of the oc- tamer are also determinants of affinity and specificity. Since the repertoire of Oct-l-binding sites has accumu- lated through many independent studies we sought to exam- ine the DNA-binding flexibility of Oct-1 without the con- straints of the gene context. We have therefore immunose- lected Oct-1 -bound random-sequence oligonucleotides using rabbit antibodies directed against a human Oct- 1 C-terminal peptide. After three rounds of enrichment, all the indepen- dently selected and sequenced clones contained quite diverse octamer-like motifs or TAATGARAT-like sequences that were capable of specific binding to Oct-1, albeit over at least a 200-fold range of affinities. Also, these Oct-l-binding se- quences were screened with a mouse whole-brain nuclear extract to demonstrate that neuronal octamer-binding pro- teins exhibit overlapping sequence specificities. Recently, Verrijzer et al. (1992) have also reported the isolation from random-sequence oligonucleotides of a collection of se- quences bound by the isolated and purified POU domain of Oct-I The similarities and differences between the two sets of data are also considered. zyxwvuts MATERIALS AND METHODS Preparation of anti- C-terminal Oct-1 peptide) serum A-20-amino-acid peptide with the sequence (NH,)- CTVASASGAASTTTTASKAQ-(COOH), corresponding to the C-terminal 19 amino acids of the human Oct-1 protein sequence, was synthesized using t-butoxycarbonyl-protected amino acids using an ABI 430A peptide synthesizer. The N- terminal cysteine was added for coupling purposes (Harlow and Lane, 1988) and the peptide was purified by reverse- phase HPLC before being coupled via its thiol group to key- hole-limpet haemocyanin (KLEI) essentially as described by Landschulz et al. (1988). Approximately 70% of the fluo- rescamine-labelled tracer peptitie was associated with KLH. Two New Zealand white rabbits were bled to obtain pre- immune Serum before intramuscular injection of 1 mg pep- tide-coupled IUH in Freund's complete adjuvant. Rabbits were boosted fortnightly using 0.5 mg or less of antigen in Freund's incomplete adjuvant and the anti-peptide serum was collected three months after the primary injection. It is re- ferred to as anti-peptide serurn to distinguish it from the polyclonal anti-Oct-I serum (Sturm et al., 1988) used in Fig. 4B that contains antibodies directed against the POU domain and which inhibits DNA binding. Oligonucleotides, probes and competitor DNA fragments The random sequence oligonucleotide 5'-GCTGGATCC- TACCAC[N,,]TCTAGATCGAGCTCG-3', forward primer 5'-CGAGCTCGATCTAGA-3' and the reverse primer 5'- GCTGGATCCTACCAC-3' were synthesized using an ABI 391 oligonucleotide synthesizer and purified according to Applied Biosystems protocols. The restriction fragment probes for gel mobility-shift assays or immunoselection con- trols were 3'-end '2P-labelled HindIII-EcoRI digests of pUCl19H2B-box' (H2B wild- ype octamer-containing frag- ment, H2B wt; Baumruker et al., 1988), pUC119B20wt' (SV40 site-I Octa-1 probc, SV40 Octa 1) and pUC119B20dpm8' (in which tao positions of the SV40 site- I Octa-1 probe had been mutated to destroy Oct-1 binding, SV40 Octa 1 dpm8; Sturm et al., 1988), BamHI-PvuII di- gest of pUCll8AO (containing both SV40 octamer-binding sites Octa 1 and Octa 2; Sturm et al., 1987) and HindIII- PstI digest of pp6xB20 (containing six copies of the SV40 enhancer Octa-1 site; Ondek et al., 1988). A 40-bp H2B ds oligonucleotide containing the chicken H2B perfect octamer, 5'-GATCCTAGCCCCTCTATGCAA- TGAGAAGCATTCCTTTCGGATC-3' (H2B oligonucleo- tide) and the HindIII-EcoRI fragment of pUC119H2B-box' were used as positive Oct-l-binding competitors. The 50-bp ds oligonucleotide 5'-CATGGKKCATGCGATGTAGGCC- ACGTGACCGGGTTCTAGATCGAGCTCG-3', containing sequence from the adenovirus-2 major late promoter (MLP oligonucleotide) and the HiiidIII-EcoRI fragment of pUC119B20dpm8+, were used zyx is non-competing DNA for Oct-1 binding. Individual selected Oct-l-binding sequences were used as competitors by employing the M13 -20 and reverse-sequencing primers to amplify a 206-bp fragment of the pBS + (Stratagene) polylinker containing the various cloned binding sites with the polymerase chain reaction (PCR). The amplified DNA fragments were gel-purified using a Mermaid kit (BIO101). Mouse brain and HeLa cell total nuclear extract Brain nuclear extracts were prepared from 0.05 g fresh mouse brain kept at 4°C prior to extraction using a slightly modified mini-extraction method (Schreiber et al., 1989) in- volving homogenization in buffer A (10 mM Hepes/KOH, pH7.9, 10mM KC1, 0.1 nM EDTA, 0.1 mM EGTA, 1 mM  801 dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride (PhMe- SO,F), 2 pg/ml aprotinin, 2 pg/ml leupeptin, 2 pg/ml peps- tatin) with a type-A dounce homogeniser (30 strokes). The debris was allowed to settle for 5 rnin before the supernatant was removed to an Eppendorf tube on ice and the extraction continued as for cultured HeLa cells. 1 zyxwv   lo6 HeLa cells were harvested by scraping, washed in 0.138 M NaCI, 3 mM KC1, 8 mM Na,HPO,, 1.2 mM KH2P04, pH 7.2 (NaCl/P,), resus- pended in 400 pl ice-cold buffer A, swollen on ice for 15 rnin and lysed by addition of nonidet P40 to 0.6 . The nuclei were pelleted, resuspended in 100 p1 buffer B (20inM HepedKOH, pH 7.9, 400 mh4 NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 1 mM PhMeSO,F, 2 pg/ml aprotinin, 2 pg/ml leupeptin, 2 pg/ml pepstatin), shaken on ice for 15 min, centrifuged for 5 min at 11 000 g and the sol- uble nuclear fraction was frozen at -70°C. zyxwvut Electrophoretic mobility-shift assays HeLa heparin garose-fractionated nuclear extract (Sturm et al., 1987), an unfractionated HeLa total nuclear extract (see above) or a mouse brain total nuclear extract was initially incubated with 1 pg poly(d1-dC) poly(d1-dC) in 12 mM Hepes, pH 7.9, 12% (by vol.) glycerol, 80 mM KCl, 0.6 mM EDTA, 0.6 mM dithiothreitol, 0.2 mM PhMeSO,F, 100 pg/ml bovine serum albumin, zyxwvuts .05 nonidet P40 at 25 C for 10 rnin before the addition of radiolabelled probe, either alone or with unlabelled competitor DNA. Specific protein-DNA complexes were allowed to form for a further 15 min in a final volume of 20 pl or 30 pl. When antiserum was present in the reaction it was added with the protein extract and the initial incubation time was extended to 30 min. Quantitative electrophoretic mobility-shijit assays Plasmid DNA was treated with EcoRI and SalI and end- labelled in the presence of a fixed amount of a longer control fragment that could also only be labelled at an EcoRI end. The fragments were separated by gel electrophoresis, re- moved and quantified. The long fragment controlled for variations in efficiency of end-labelling associated with indi- vidual minipreps. 20 fmol of the test probe was mixed with 5 pg (1 pl) or 10 pg heparin-agarose fraction and binding reactions were performed for 15 min as above. The samples were analysed by electrophoretic mobility-shift assay (EMSA). The shifted complexes and free DNA were re- moved from the gel and the Cerenkov radioactivity was mea- sured to determine the proportion of input DNA bound by the protein. Oct-1-binding-site selection Binding reactions The binding reaction performed as described above con- tained 5 pg HeLa heparin - garose-fractionated nuclear ex- tract and 1 pl anti-peptide serum. 5 fmol gel-purified ds ran- dom oligonucleotide, that had been radiolabelled by zyxw a 32P]dATP and [a-32P]dCTP incorporation during the exten- sion of the forward primer, was added after a 30-niin initial incubation at 25 C. Protein-DNA complexes were allowed to form for a further 15 min in a final reaction volume of 20 pl. Immunoprecipitation The recovery of Oct-1 -bound oligonucleotides was essen- tially as described by Pollock and Treisman (1990) except that a buffer containing 20 mM Hepes, pH 7.9, 100 mM KCI, 20% (by vol.) glycerol, 1 mM EDTA, 1 mM dithiothreitol and 0.2 mM PhMeSO,F was used for swelling and washing the protein-A- Sepharose. Probe amplification The recovered DNA was amplified in 50 1-11Tay polymer- ase buffer (Promega] containing 20 pmol phosphorylated for- ward and reverse primers, 200 pM each dGTP and dTTP, 50 pM each dATP and dCTP, 10 pCi each [a-'2P]dATP and [a- PIdCTP (3000Ci/mmol, Bresatec) and 2 U Taq polymer- ase (Promega). The amplification program was 94°C for 1 min, 55°C for 1 min and 72°C for 1 rnin for 20 cycles in a Corbett FTS-1 thermal cycler. The product was gel puri- fied, eluted and quantified by measuring the Cerenkov radio- activity. The selection protocol was repeated twice. Cloning The gel-purified oligonucleotide was treated with T4 DNA polymerase (Toyobo) in 37.5 mM Trishcetate, 150 mM potassium glutamate, 15 mM magnesium acetate, 0.75 mM 2-mercaptoethanol and 100 pM dNTPs for 30 min at 15°C to ensure that PCR-amplified ends were blunt before cloning into the SmaI site of pBS+. The plasmids were se- quenced using a Sequenase lut (USB). RESULTS Anti- C-terminal Oct-1 peptide) serum specifically recognises an Oct-1-DNA complex and enriches a population of random-sequence oligonucleotides for Oct-1-binding sites To separate oligonucleotides specifically bound to Oct-1 from the milieu of protein-bound and free oligonucleotides we have used antibodies directed against a C-terminal Oct-1 peptide which allow the selective adsorption of Oct-l-oligo- nucleotide complexes to protein-A - epharose. Western-blot analysis of a heparin- agarose-fractionated HeLa nuclear ex- tract demonstrated that the anti-peptide serum specifically recognised a single band at approximately 100 kDa, which is the expected size for Oct-1, a band at approximately 40 kDa and other minor bands that cross-reacted with endogenous rabbit antibodies in the pre-immune serum (data not shown). The anti-peptide serum also recognised DNA-bound Oct-1 and supershifted the complex in an EMSA without signifi- cantly disrupting the binding capacity of Oct-1 (Fig. 1 A, compare lanes 2 and 4). We tested the specificity of the selection system by cap- ture of the antibody-Oct-1 -DNA complexes on protein- A- Sepharose using oligonucleotides of defined sequence. 30-40% of 5 fmol end-filled H2B wt probe was reproduci- bly recovered when 1 p1 of heparin - garose fraction and im- mune serum were present but not when the pre-immune se- rum was used or when the HeLa extract was omitted. The substitution of the SV40 Octa 1 dpm8 probe or the MLP oligonucleotide for H2B wt resulted in background retention under all conditions, in the order of 0.3-0.5% for 5 fmol input oligonucleotide (data not shown).  802 Fig. 1. Characterisation of anti- C-terminal Oct-1) serum and sequential enrichment of Oct-1-binding sites. A) zyxwvu   radiolabelled promoter fragment from the chicken H2B gene (H2B wt; lane 1) was incubated with 1 pl Oct-I-enriched HeLa heparin -agarose frac- tion either alone (lane 2), or in combination with pre-immune serum (lane 3), serum from the same rabbit after immunization with a C- terminal Oct-1 peptide (lane 4), or serum and an excess of the anti- genic peptide (lane zyxwvutsrqp ). The slight shift of the Oct-1 complex ob- served in the presence of the pre-immune serum (lane 3) was a common phenomenon and has been previously reported (Sturm et al., 1991). (B) Sequential enrichment of Oct-1-binding sites from random sequence oligonucleotides. 1 p1 heparin- agarose fraction was incubated with uniformly labelled, unselected random oligonu- cleotide (lane 1) or the oligonucleotide after one, two and three rounds of Oct-1-binding site selection (lanes 2-4, respectively). The enriched complex formed on the oligonucleotide population after three rounds of selection was challenged with a tenfold molar excess of an unlabelled H2B oligonucleotide (lane 5), or unselected ran- dom-sequence oligonucleotide (Oligo B, lane 6) or 1 pl anti-peptide serum (lane 7). Ah, antibody; H-A, eparin-agaroae fraction. Fig.2. Oct-1 complexes formed on a selection of individual im- munoselected binding sites. 1 p1 heparin-agarose fraction was in- cubated with 20 fmol 3'-end-]abelled EcoRI-Sun fragment of the clone indicated by the numbers for each lane. A population of random 20 -residue oligonucleotides that were flanked by 15-bp arms zyx f defined sequence was used as the target for selecting Oct-1 -binding sequences. Assum- ing that the 20-bp sequence was truly random then a perfect octamer would be expected to occur once in approximately 5000 molecules. Since 1 p1 heparin-agarose fraction could bind up to 2 fmol of a specific octamer-containing probe such a sample should contain sufficient protein to bind all high- affinity and low-affinity sequences present in the 20 bp ran- dom sequence, at least in the first round of selection. 5 fmol ds random-sequence oligonucleotide was mixed with the heparin-agarose fraction in a binding reaction which also contained sufficient antibody to complex at least 95% of the DNA-bound Oct-1 (Fig. 1 A, lanes 2 and 4) and mediate ad- sorption to protein-A- Sepharose. The resin-bound oligonu- cleotides were recovered, cycled through the PCR procedure and the amplified 50-bp DNA was gel purified. This enriched and amplified population was further selected until no addi- tional enrichment was observed (a total of three cycles; Fig. 1 B, lanes 1-4). In the first round of selection, 0.7% of the unselected oligonucleotide was recovered while in rounds 2 and 3, 17% and lo%, respectively, of 5 fmol amplified selected target DNA was recovered. The binding-sequence specificity of the enriched complex was confirmed by effi- ciently competing formation of the retarded complex with an excess of cold H2B oligonucleotide that contains a perfect octamer (lane 5) but not by an equivalent excess of unse- lected random oligonucleotide (lane 6). The retarded band was also supershifted by the peptide serum, confirming that the enriched complex contained Oct-1 (lane 7). Qualitative gel-shift analysis on all clones that contained a single 50-bp insert demonstrated that all formed complexes co-migrated with Oct-1 complexes; only two clones, c1.19 (Fig. 2) and c1.33 showed additional unrelated bands (data not shown). This further demonstrates that the antibody used has not se- lected for DNA sequences other than those which can be bound by Oct-1. Sequence analysis of immunoselected Oct-1-binding sites The sequences of 49 insert- containing colonies, repre- senting independently selected Oct-1-binding sites, are shown in Table 1. These sequences are generally AT-rich with an overall composition 01 69% AT, while nine se- quences have 80-90% AT content. Many of these sequences contain significant identities to known Oct-1 -binding sites in  803 Table 1. Selected binding sites for Oct-1 complexes. 49 cloned binding sites representing independently selected Oct-1-binding sequences generated by three rounds of immuno-selection using anti-(C-terminal Oct-1 peptide) serum are shown, aligned to reflect a progression in this population from a perfect octamer to highly degenerate octamer-like sequences. The numbering system in column 1 corresponds to the clone number where the suffix .I zyxwvutsrq r .2 indicates that the sequence comes from a double insert derived during blunt-end cloning. Sequences that are underlined indicate matches to previously defined Oct-1 -binding sites. The extent of the match is shown in the third column and a description of the site plus reference is shown in the fourth and fifth columns. ch, chicken; hu, human; mu, murine; Ad2, adenovirus 2; MMTV, mouse mammary tumour virus (ITR, inverted terminal repeats). Sequences contributed by either the defined sequence arm of oligonucleotide B are shown in lower case. Note that for two of these sequences the best match is not to an octamer but instead to a TAATGARAT sequence. Clone Sequence no. Number of identical Naturally occurring Reference residues compared to site naturally occurring site 4 7 45 21 22.1 41.2 26 39 29 2 6 16.2 9 38 15.1 17 30.2 3 8 12 18 20 15.2 19 30.1 33 40 44 11 28 34 13 23 22.2 1 31 14 35 36 37 41.1 16.1 42 41 48 46 32 25 21 zyxwvutsr G GC TGC T mT TT T TC G TGC T TT gaT CGC T T TCTC T TT GT TTC T GTTTTGT CCTTTT TTC TTC T gt T CCCC T TTC T CT TT GTTTCG E TGT T gtg TG T TGG T CmTGG zyxwvu   gaT TGCT T C C TGC T G TG TT TGC T T C GTTGG TG T TGC T t cta T GTGCGT TT TGC G t cta GTC TTC TGC G T Cg CTC TG T TGC GTBGC T CC T G GCT TGC q tgg T TGBTT TGC GT GC CGC T TGC C TC C TTG C TTGC T TCCC aga GTC T C G GGGTG GG GGGT TGCT G CGT GC TCGTGC T GGC gt T C TGT G T TWCC GC aTT CGC G GT T TGCG TG GT TCC T TTG TTT T TT TTC T T T TT C CGGTTG GC GZ TG T t cta CTG TGT T T TC TGC C TGC TGGC T CTG CT TGC G C T TGCTG T TGC T TTG TG TGCT CT CCC TCCC GGT T T TTC TTG TT TGGT TTGTG TT T CT T TG CGT T C CCC CT C T T T T T CCTG CCG CTTT T tc TTT TG CT T T TG C cac GT T T T GGCGGTTTCG TT CGC T TC GGTC TC GT T GGGT T TTT G GT GT GT T T zyxwvu  t TT T TTC G GCCTCT aga CTGT Tm TCGT cG TGTC G G T G TT TT TGTTT T TC GGGT C G G TG T TC T CCTC T TG TT TT GCG TG GGTT TT T TCGCGGT TT G T TG GTT G GT T GC T T T TG TT GTG 11/11 11/12 10/11 11/11 13/15 1011 1 919 zyxwvu 18 919 17/20 919 10/13 loll1 911 0 911 0 11/12 1011 1 10112 1011 2 1411 6 ch H2B promoter hu U2 snRNA promoter Baunmker et al., 1988 Baumruker et a]., 1988 hu IgH distal enhancer MMTV promoter, site D hu U3 snRNA enhanced U3 zyxw ox SV40 enhancer, Octa 2 Ad2 ITR mu U1 b snRNA mu U1 b snRNA hu VH H63 heptamer-octamer SV40 enhancer, Octa 1 7SK RNA promoter, medial site Schreiber et a]., 1989 Briiggemeier et a]., 1991 Suh et al., 1986 Sturm et al., 1987 Pruijn et al., 1987 Lea et al., 1991 Lea et al., 1991 Falkner and Zachau, 1984 Sturm et al., 1987 Murphy et al., 1989 SV40 enhancer, Octa 1 Sturm et al., 1987 hu U1 snRNA promoter Drosophila zyxw elanogaster abd-B gene rat prolactin promoter Skuzeski et al., 1984 Schreiber et a]., 1989 Voss et a]., 1991 hu -globin promoter Ponce et al., 1991 hu c-myc promoter hu IL-2 promoter, proximal site Tahmoto et al., 1989 Felli et al., 1991 hu U3 snRNA enhancerm3 box TAATGARAT TA ATG AR AT Suh et al., 1986
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks